Abstract: To solve defects such as volume shrinkage and warping deformation that occur during injection molding process of automobile B-pillar upper trim panels, injection molding process of talc-modified polypropylene material was simulated using Moldex3D software. Sixteen orthogonal experiments were designed with injection time, melt temperature, mold temperature, holding pressure, holding time as variables, volume shrinkage and Z-direction (product demolding direction) warpage as objectives. Critic weighting method was used to calculate weights of these two parameters, and multi-objective optimization was transformed into single-objective optimization through comprehensive scoring. Finally, range analysis of comprehensive scores revealed following order of influence of five process parameters: holding pressure > mold temperature > filling time > melt temperature > holding time. Optimal molding process parameter combination was: filling time 2 s, melt temperature 240 ℃, mold temperature 30 ℃, holding pressure 70 MPa, and holding time 8 s. Simulation of optimal molding process parameter combination yielded a volume shrinkage rate of 5.901% and a Z-direction warpage of 1.75 mm. Compared with initial analysis results, volume shrinkage rate decreased by 12.2%, and Z-direction warpage decreased by 9.04%. Actual mold trials verified that product was fully filled, of good quality, and met production requirements.
With steady advancement of automotive lightweighting in recent years and continuous development of China's plastics industry, application categories and scope of plastics in automotive industry are also gradually expanding, interior and exterior trim parts of automobiles have basically achieved plasticization. Automotive pillar trim panel is an indispensable part of automotive interior trim. On the surface, it can not only enhance appearance of interior of car, but also protect surface of pillar, avoid wear and scratches in daily use, hide connecting parts and wiring harness inside pillar, improving comfort of interior of car; structurally, it can enhance structural stability of pillar, play an important role in connecting top and bottom of car body, making car body more stable and strong, can also play a role in partial buffering and energy absorption to reduce impact force on occupants during collision.
In actual injection molding production process, there are many factors that affect molding quality of such plastic parts, especially in selection of process parameters. Inappropriate molding process parameters can cause molding defects such as warping, flash, and weld lines. There have been many studies on optimization of injection molding process parameters. Sun Xiaoxia et al. took refrigerator drawers as research object, used progressive orthogonal experiments and combined grey relational analysis to quickly find the best process molding parameters, which improved product quality. Li Shu et al. used entropy weight method to optimize warping deformation and volume shrinkage defects of left front bumper of automobiles, and obtained optimal process parameters through range analysis of comprehensive scores. Lo used Moldflow to simulate injection molding process of computer cooling fan impeller, used Taguchi method and grey relational method to determine the best process parameters to solve warping problem, so as to reduce trial molding cost and improve product quality. Zhu Hongping et al. used theoretical analysis, finite element simulation and experimental verification to analyze molding quality of washing machine water box cover, and used weld line as optimization target to conduct Taguchi experiment, so as to obtain optimal process combination, verified its molding quality through mold design and trial molding. Hiyane-Nashiro et al. proposed EAAWSM weighted search method. By comparing it with three optimization methods, Taguchi-Gray, TOPSIS, and MOGA, it effectively reduced shrinkage rate and warpage of plastic parts, indicating that this method has reliability in optimizing advantages and results of two or more variables. Ren Lihui et al. used warpage deformation, shrinkage index, and volume shrinkage rate as evaluation indicators to improve injection molding quality of toilet chair panels. They used Critic method to determine weight coefficients of each evaluation indicator and adopted grey relational comprehensive evaluation method based on TOPSIS to obtain optimal combination of injection molding process parameters for plastic parts. Most of above studies have carried out process optimization through complex calculations or programs, which is time-consuming and labor-intensive in production and cannot guarantee improved production efficiency and cost savings. Advantage of using Critic weighted method to optimize process parameters is that it can effectively handle multi-objective problems, find a relatively balanced solution, adjust weights according to actual needs to accurately reflect importance of objectives and obtain more reasonable process parameters; calculation speed is relatively fast, which can improve production efficiency, reduce energy consumption, and reduce production costs. Based on above analysis, this paper takes B-pillar trim panel of an automobile as research object and uses Moldex3D software to simulate its injection molding process. Based on initial process analysis, orthogonal experiments are designed, weights of volume shrinkage rate and Z-direction (product demolding direction) warpage deformation are calculated using Critic weighting method. Then, through range analysis of comprehensive score, multi-objective optimization is transformed into single-objective optimization to obtain optimal molding process parameters. Finally, simulation and actual mold testing are combined to improve product quality and yield.

A three-dimensional model of a certain model of automobile B-pillar trim panel is established using UG10.0. Length of a single plastic part is 445 mm, width is 252 mm, height is 60 mm, volume is 323,049.73 mm³, and average wall thickness is approximately 2.5 mm. Figure 1 shows three-dimensional model of automobile B-pillar trim panel. As can be seen from Figure 1, back of plastic part has a complex structure, including multiple clips and reinforcing ribs. Molding process of this plastic part requires that surface be smooth and clean, free of flash, burrs, weld lines and other defects, and that it be able to be correctly installed with other parts. Therefore, large warping and volume shrinkage should be avoided during molding process.

 

Fig. 1 Three-dimensional model of trim panel on B-pillar of a car

We selected polypropylene material with model P221T-UV containing 15% talc filler, produced by Suzhou Xuguang Polymer Co., Ltd. This material is often used in molding automotive interior parts, has good mechanical properties and excellent UV aging resistance. Recommended process parameters for this material are listed in Table 1.

Table 1 Recommended process parameters for PP/15%talc material

In order to quickly fill cavity with molten material, save raw materials, reduce pressure and heat loss, a composite feeding structure combining hot runner and ordinary runner is adopted. Plastic part has high requirements for surface quality, so a submarine horn gate is used for one-mold two-cavity molding. Gating system scheme is shown in Figure 2.

 

Figure 2 Casting system

First, plastic part model is repaired and simplified in CAD doctor. Repaired and simplified model is imported into Moldex3D to perform mesh division on plastic part and runner system respectively, manually repair mesh defects. Final total number of meshes is 1,487,491. Figure 3 shows mesh division result and number of trim panel on B-pillar of car.

 

Fig. 3 Mesh delineation results and number of trim plates on B-pillar of car

In Moldex3D, initial process parameters for simulating injection molding of this plastic part were set as follows: filling time 3 s, plastic temperature 220 ℃, mold temperature 45 ℃, holding time 10 s, and holding pressure 70 MPa. An analysis of "filling + holding + warpage" was performed on initial process parameters. Figure 4 shows initial simulation results. Volume shrinkage rate of this plastic part was 6.721%, Z-direction warpage deformation was 1.924 mm upward and 1.445 mm downward. Uneven volume shrinkage and excessive Z-direction warpage can significantly affect product, such as causing dimensional instability and mismatch during installation. Therefore, both need to be controlled within range of 6.3% and 1.8 mm respectively. Initial analysis results did not meet requirements. Therefore, further optimization of volumetric shrinkage rate and Z-direction warpage was conducted to meet design specifications, improve part quality, and increase production efficiency.

 

Fig. 4 Initial Analysis

A uniform volumetric shrinkage rate and low Z-direction warpage depend on adjustment of process parameters during molding process. Therefore, selecting appropriate molding process parameters is crucial. Injection time (A), melt temperature (B), mold temperature (C), holding pressure (D), and holding time (E) were selected as variables, with volumetric shrinkage rate and Z-direction warpage as optimization objectives. Based on molding material parameters used, four levels were selected for orthogonal experimental design. Orthogonal experimental factor levels are shown in Table 2.

 

Table 2 Orthogonal test factor level table

According to factor level table in Table 2, it is necessary to select L16 (45) orthogonal table for test. Moldex3D mold flow analysis software was used to simulate 16 injection molding processes, obtain warping deformation and volume shrinkage rate of B-pillar trim panel body in Z direction for each test. Results of orthogonal test are shown in Table 3.

 

Table 3 Results of orthogonal test

Critic weight method is to convert multi-objective optimization into single-objective optimization by calculating comprehensive score according to weight of each objective, finally perform range analysis to obtain main and secondary factors affecting objective and optimal molding scheme. Objective of this test is to simultaneously optimize two defects of volume shrinkage rate and Z-direction warping deformation in molding of B-pillar trim panel of car, and to seek optimal molding process scheme for both by changing process parameters. Since volume shrinkage rate and Z-direction warpage to be optimized have different dimensions and units, they must first be dimensionless to make comparison and analysis of data more convenient and accurate. Calculation formula is shown in Equation (1).

 

(1) Where:  represents dimensionless value;  represents experimental value corresponding to index j in i-th experiment; minxij represents minimum value in this experiment; maxxij represents maximum value in this experiment.
In Critic weighting method, in order to reflect dispersion of a dataset, standard deviation is used to represent difference and fluctuation of values within each index. The larger standard deviation, the greater numerical difference of index, the more information it can reflect, and the stronger evaluation intensity of index itself. Calculation formula is Equation (2).

 

(2) Where: Sj represents standard deviation of j-th index;  represents average value of j-th index; n represents number of experiments, n is 16.
Information content represents role of evaluation index in the entire evaluation index system. Its calculation formula is Equation (3).

 

(3) Where: Cj represents information content of j-th indicator; rij represents correlation coefficient between evaluation indicators i and j.
Weight refers to percentage of a certain factor or indicator, representing relative importance of that factor or indicator in the overall evaluation, and its calculation formula is formula (4).

 

Where:  represents weight of j-th indicator.
Critic comprehensive score is calculated as shown in formula (5):

 

Where: I represents comprehensive score.
Substituting experimental results in Table 3 into formulas (1)~(5), values of dimensionless, standard deviation, information content, weight, and comprehensive score are listed in Table 4.

 

Table 4 Calculated data for each parameter
Final comprehensive score value is analyzed by range analysis, as shown in Table 5. The larger range, the greater impact of that factor on the overall score.

 

Table 5 Composite score extreme variance analysis
Analysis of range of composite scores in Table 5 shows that holding pressure has the greatest impact on the overall score, followed by mold temperature, filling time, and melt temperature, while holding time has least impact. Therefore, optimal molding process parameter combination for the overall score is A1B4C1D3E2, i.e., filling time 2 s, melt temperature 240 ℃, mold temperature 30 ℃, holding pressure 70 MPa, and holding time 8 s.

Based on above optimized molding process parameter combination A1B4C1D3E2, simulation analysis was performed in Moldex3D. Simulation results of optimized process parameter combination for the overall score are shown in Figure 5. A comparison of initial process parameter combination and optimized molding parameter combination for the overall score is shown in Table 6.

 

Fig. 5 Optimal process parameter analysis results

 

Table 6 Comparison of initial and optimal process parameters
As shown in Figure 5 and Table 6, optimal combination of process parameters, after comprehensive scoring optimization, resulted in a volume shrinkage rate of 5.901% and Z-direction warpage of 1.75 mm upwards and 1.289 mm downwards, representing reductions of 1.466% and 0.174 mm respectively compared to initial molding analysis. These reductions represent decreases of 12.2% and 9.04%, respectively, yielding relatively superior molding results and significantly improving quality of molded part.
To verify accuracy of experiment, optimal process parameter combination A1B4C1D3E2 was input into injection molding machine for on-site mold testing. Molded sample is shown in Figure 6. Observation of molded sample showed that molding effect of plastic part was good, with no obvious surface defects. Furthermore, its volume shrinkage rate and Z-direction warpage met requirements, satisfying normal assembly. This verifies feasibility of optimization method combining orthogonal experiments and Critic weighting method, improving both product quality and production efficiency.

 

Fig. 6 Automotive B-pillar upper trim plate molding

(1) Based on Moldex3D mold flow analysis software, volume shrinkage rate and Z-direction warping deformation of automotive B-pillar upper trim plate were analyzed. An orthogonal experiment combined with Critic weighting method was used to convert two defects into a comprehensive score value for process parameter optimization.
(2) By calculating range of comprehensive score, order of factors affecting two defects was determined as: holding pressure > mold temperature > filling time > melt temperature > holding time. Optimal molding process scheme A1B4C1D3E2 was finally obtained, namely, filling time 2 s, melt temperature 240 ℃, mold temperature 30 ℃, holding pressure 70 MPa, and holding time 8 s.
(3) Through simulation analysis of optimal combination of process parameters, volume shrinkage rate was found to be 5.901% and warping deformation in Z direction was 1.75 mm. Compared with initial analysis, two decreased by 12.2% and 9.04% respectively. Combined with on-site mold trial verification, molding quality of plastic part is good, meets requirements of normal assembly and mass production.